Selected Papers by Faculty in the Department of Biostatistics

Warning: Summaries were generated by Copilot, and may contain errors.

Emine Bayman

  • CV
  • Statistical Methodologies: Emine Bayman focuses on Bayesian methods, including Bayesian clinical trial design, Bayesian outlier detection, and Bayesian hierarchical modeling. She also works on cluster randomized trials, modified loss functions, and statistical modeling for clinical metrics and pain trajectories.
  • Interdisciplinary Applications: Her interdisciplinary research spans pain science, anesthesiology, clinical trials, neurology, fibromyalgia, opioid use, and brain development in children. She collaborates extensively on studies involving post-surgical pain, chronic pain prediction, and behavioral interventions in neurological and musculoskeletal conditions.
  1. Bayman EO, Chaloner K, Cowles MK. Detecting qualitative interaction: a Bayesian approach. Statistics in medicine. 2010;29(4):455-463. Epub 2009/12/02. doi: 10.1002/sim.3787. PubMed PMID: 19950107
  2. Bayman EO, Chaloner KM, Hindman BJ, Todd MM, IHAST I. Bayesian methods to determine performance differences and to quantify variability among centers in multi-center trials: the IHAST trial. BMC medical research methodology. 2013;13:5. Epub 2013/01/18. doi: 10.1186/1471-2288-13-5. PubMed PMID: 23324207; PMCID: 3599203
  3. Bayman EO, Dexter F, Todd MM. Assessing and Comparing Anesthesiologists’ Performance on Mandated Metrics Using a Bayesian Approach. Anesthesiology. 2015;123(1):101-115. Epub 2015/04/24. doi: 10.1097/aln.0000000000000667. PubMed PMID: 25906338
  4. Bayman EO, Dexter F, Todd MM. Prolonged Operative Time to Extubation Is Not a Useful Metric for Comparing the Performance of Individual Anesthesia Providers. Anesthesiology. 2015;124(2):322-338. Epub 2015/11/07. doi: 10.1097/aln.0000000000000920. PubMed PMID: 26545101
  5. Bayman EO, Parekh KR, Keech J, Selte A, Brennan TJ. A Prospective Study of Chronic Pain after Thoracic Surgery. Anesthesiology. 2017;126(5):938-951. doi: 10.1097/ALN.0000000000001576. PubMed PMID: 28248713; PMCID: PMC5395336

Patrick Breheny

  • CV
  • Statistical Methodologies: Patrick Breheny develops and applies methods for high-dimensional data analysis, penalized regression, variable selection, false discovery rate control, and computational statistics. He has contributed to software development in R for penalized models (e.g., ncvreg, grpreg, biglasso, visreg) and works on penalized linear mixed models, nonconvex optimization, and bi-level variable selection.
  • Interdisciplinary Applications: His collaborative research spans genomics, genetic epidemiology, cancer biology, neuroendocrine tumors, pregnancy and birth outcomes, infectious disease, and neuroscience. He has worked extensively on gene expression analysis, copy number variation, olfactory receptor studies, and clinical outcomes in oncology and maternal health.
  1. Breheny P (2018). Marginal false discovery rates for penalized regression models. Biostatistics, 20: 299-314.
  2. Breheny P (2015). The group exponential lasso for bi-level variable selection. Biometrics, 71: 731–740.
  3. Breheny P and Huang J (2015). Group descent algorithms for nonconvex penalized linear and logistic regression models with grouped predictors. Statistics and Computing, 25: 173-187.
  4. Breheny P and Huang J (2011). Coordinate descent algorithms for nonconvex penalized regression, with applications to biological feature selection. Annals of Applied Statistics, 5: 232–253.
  5. Breheny P and Burchett W (2017). Visualization of regression models using visreg. The R Journal, 9: 56–71.

Grant Brown

  • CV
  • Statistical Methodologies: Grant Brown focuses on Bayesian computing, statistical learning, models of dynamic processes, spatiotemporal modeling, compartmental epidemic models, approximate Bayesian computation (ABC), and correlated data analysis. He has developed software tools for Bayesian epidemic modeling and effect visualization in black-box models.
  • Interdisciplinary Applications: His applied research spans infectious disease ecology (e.g., visceral leishmaniasis, Lyme disease, COVID-19, Ebola), immunological modeling, stroke triage, substance use disorders, environmental health, speech and hearing sciences, and public health policy. He collaborates across disciplines including epidemiology, nursing, engineering, and behavioral health.
  1. Phillip K, Nair N, Kamalika S, Azevedo JF, Brown GD, Petersen CA, Gomes-Solecki M. (2021). Maternal transfer of neutralizing antibodies to B. burgdorferi OspA after oral vaccination of the rodent reservoir. Vaccine. DOI: 10.1016/j.vaccine.2021.06.025
  2. Seedorff N, Brown G D (2021). totalvis: A Principal Components Approach to Visualizing Total Effects in Black Box Models. SN Computer Science. DOI: 10.1007/s42979-021-00560-5
  3. Brown GD, Oleson JJ, Porter AT (2016). An empirically adjusted approach to reproductive number estimation for stochastic compartmental models: A case study of two Ebola outbreaks. Biometrics. DOI: 10.1111/biom.12432
  4. Brown GD, Porter AT, Oleson JJ, Hinman JA, (2018). Approximate Bayesian computation for spatial SEIR(S) epidemic models. Spatial and Spatiotemporal Epidemiology. DOI: 10.1016/j.sste.2017.11.001
  5. Ozanne MV, Brown GD, Toepp AJ, Scorza BM, Oleson JJ, Wilson ME, Petersen CA (2020). Bayesian Compartmental Models and Associated Reproductive Numbers for an Infection with Multiple Transmission Modes. Biometrics. DOI: 10.1111/biom.13192

Joe Cavanaugh

  • CV
  • Statistical Methodologies: Joseph Cavanaugh specializes in model selection, variable selection, time series analysis, state-space modeling, information criteria (e.g., AIC, BIC), and model diagnostics. He has developed numerous discrepancy-based and bootstrap-based approaches for model comparison and selection, and contributed extensively to the theory and application of Kullback-Leibler divergence in statistical inference.
  • Interdisciplinary Applications: His applied research spans epidemiology, infectious diseases, injury prevention, clinical decision-making, public health policy, dental health, and transportation safety. He collaborates on projects involving cystic fibrosis, sepsis, bullying prevention, motor vehicle crashes, and agricultural health, often using statistical modeling to inform policy and clinical practice.
  1. Riedle B, Neath AA, and Cavanaugh JE (2020). Reconceptualizing the p-value from a likelihood ratio test: a probabilistic pairwise comparison of models based on Kullback-Leibler discrepancy measures, Journal of Applied Statistics. DOI: 10.1080/02664763.2020.1754360.
  2. Cavanaugh JE, Neath AA (2019). The Akaike information criterion: Background, derivation, properties, application, interpretation, and refinements. WIREs Computational Statistics, 11:e1460. DOI: 10.1002/wics.1460.
  3. Peterson RA, Cavanaugh JE (2019). Ordered quantile normalization: A semiparametric transformation built for the cross-validation era. Journal of Applied Statistics. DOI: 10.1080/02664763.2019.1630372.
  4. Zhang T, Cavanaugh JE (2016). A multistage algorithm for best–subset model selection based on the Kullback–Leibler discrepancy. Computational Statistics, 31(2):643-669. DOI: 10.1007/s00180-015-0584-8.
  5. Yang M, Cavanaugh JE, Zamba GJ (2015). State-space models for count time series with excess zeros. Statistical Modelling, 15(1):70-90. DOI: 10.1177/1471082X14535530.

Chris Coffey

  • CV
  • Statistical Methodologies: Christopher Coffey specializes in clinical trial methodology, particularly adaptive designs, internal pilot studies, sample size re-estimation, interim monitoring, and biomarker validation. He has extensive experience in Bayesian and frequentist methods, longitudinal data analysis, and survival analysis. He has also contributed to the development of statistical software and tools for trial design.
  • Interdisciplinary Applications: His interdisciplinary work spans neurology, neuroscience, Parkinson’s disease, multiple sclerosis, migraine, stroke, ALS, Huntington’s disease, pain research, and cardiovascular and metabolic diseases. He leads and collaborates on large-scale clinical trials and biomarker studies, including the Parkinson’s Progression Markers Initiative, NeuroNEXT, and the Acute to Chronic Pain Signatures (A2CPS) program.

Jeff Dawson

  • CV
  • Statistical Methodologies: Jeffrey Dawson works extensively on longitudinal data analysis, repeated measures, summary statistics, survival analysis, time-series modeling, and ordinal regression models. He has also contributed to clinical trial design, missing data methods, and statistical ethics and reproducible research.
  • Interdisciplinary Applications: His interdisciplinary research spans public health, neurology, cardiovascular disease, driving safety and aging, Parkinson’s and Alzheimer’s disease, sleep disorders, pediatric development, and infectious diseases. He has collaborated on numerous studies involving driver behavior, colon cancer screening, aerobic exercise interventions, and neuroimaging in neurological and developmental disorders.

Jake Oleson

  • CV
  • Statistical Methodologies: Jacob Oleson specializes in Bayesian methods, hierarchical modeling, spatial and spatio-temporal statistics, small area estimation, survey statistics, longitudinal data analysis, and statistical computing. He has developed and applied Bayesian compartmental models, hierarchical growth curve models, spatio-temporal epidemic models, and methods for analyzing ecological momentary assessment data. His work also includes methodological innovations in mixed models, functional data analysis, and statistical approaches for speech and hearing sciences.
  • Interdisciplinary Applications: His research spans audiology, otolaryngology, speech-language pathology, developmental language disorders, public health, cancer epidemiology, infectious disease modeling (e.g., visceral leishmaniasis, Lyme disease, COVID-19), environmental health, pediatric hearing loss, and cognitive neuroscience. He collaborates extensively on projects involving cochlear implants, hearing aid technologies, language development in children, spatial disease mapping, and health disparities in rural populations.
  1. Kliethermes SA, Oleson JJ. A Bayesian approach to functional mixed effect modeling with binomial outcomes. Statistics in Medicine, 33(18):3130-3146, 2014
  2. VanBuren J, Oleson JJ, Zamba GKD, Wall M. Integrating independent spatio-temporal replications to assess population trends in disease spread. Statistics in Medicine. 35(28):5210-5221, 2016. PMID: 27453437
  3. Seedorff M, Oleson JJ, McMurray B. Detecting when timeseries differ: Using the Bootstrapped Differences of Timeseries (BDOTS) to analyze visual world paradigm data (and more). Journal of Memory and Language, 102:55-67, 2018.
  4. Zahrieh D, Oleson JJ, Romitti PA. Quantifying geographic regions of excess stillbirth risk in the presence of spatial and spatio-temporal heterogeneity. Spatial and Spatio-Temporal Epidemiology. 29, 97-109, 2019.
  5. Ozanne M, Brown G, Toepp A, Scorza B, Oleson J, Wilson M, Petersen C. Bayesian compartmental models and associated reproductive numbers for an infection with multiple transmission models. Biometrics. (early view published online) 2020.

Emily Roberts

  • CV
  • Statistical Methodologies: Emily Roberts works on causal inference, Bayesian methods, clinical trial design, longitudinal data analysis, survival analysis, and small-area estimation. She has developed methods for surrogate endpoint validation, including illness-death frailty models and principal stratification, and has created several R packages and Shiny apps for applied causal analysis.
  • Interdisciplinary Applications: Her interdisciplinary research spans oncology, diabetes, mental health and suicide prevention, environmental epidemiology, microbiome studies, and auditory health. She collaborates on projects involving cancer survival, telomere biology, racial disparities, and clinical outcomes in liver transplantation and cochlear implant users.

Dan Sewell

  • CV
  • Statistical Methodologies: Daniel Sewell specializes in social network analysis, Bayesian methods, clustering, Monte Carlo methods, statistical computing, and data visualization. He has developed and applied latent space models, edge clustering techniques, and hierarchical Bayesian clustering, particularly for dynamic and longitudinal data.
  • Interdisciplinary Applications: His work spans infectious disease epidemiology (e.g., Clostridioides difficile, enteric pathogens), public health, healthcare systems, environmental health, mental health, and policy analysis. He collaborates extensively on projects involving Kenyan urban health, COVID-19 modeling, Huntington’s disease, and healthcare communication networks.
  1. Sewell DK (2020). Model-based edge clustering. Journal of Computational and Graphical Statistics, 30(2):390-405.
  2. Sewell DK, Baker, KK (2025). Estimating Risk Factors for Pathogenic Dose AccrualFrom Longitudinal Data. Statistics in Medicine, 44(23-34):e70291.
  3. Sewell, D. (2024). Posterior shrinkage towards linear subspaces. Bayesian Analysis, 1 (1), 1–24.
  4. Jang H, Justice S, Polgreen PM, Segre AM, Sewell DK, Pemmaraju SV (2019). Evaluating architectural changes to reduce infection spread in a dialysis unit. International Conference on Advances in Social Networks Analysis and Mining ’19
  5. Medgyesi, D., Sewell, D. K., Senesac, R., Cumming, O., Mumma, J., & Baker, K. K. (2019). The landscape of enteric pathogen exposure for children during play in public domains of low-income, kisumu, kenya. PLOS Neglected Tropical Diseases, 13 (3), e0007292.

Brian Smith

  • CV
  • Statistical Methodologies: Brian Smith works extensively in Bayesian methods, hierarchical modeling, spatial statistics, statistical computing, MCMC, machine learning, and quantitative imaging biomarker modeling. He has developed numerous statistical software packages and tools for Bayesian analysis, diagnostic imaging studies, and clinical trial design.
  • Interdisciplinary Applications: His interdisciplinary research is focused on cancer research, including clinical trials, epidemiology, radiomics, quantitative medical imaging, and environmental health (e.g., radon exposure). He collaborates widely across oncology, radiology, and public health, with applications in glioblastoma, pancreatic cancer, lymphoma, and lung cancer, among others.

Kai Wang

  • CV
  • Statistical Methodologies: Kai Wang works extensively in statistical genetics, genetic epidemiology, causal inference (especially Mendelian randomization), mediation analysis, bioinformatics, and deep learning. His methodological contributions include Bayesian approaches, score statistics, penalized regression, kernel association tests, and variance component models for genome-wide association studies.
  • Interdisciplinary Applications: His interdisciplinary research spans genomics, ophthalmology, environmental health, neurodevelopmental disorders, autism, glaucoma, macular degeneration, cystic fibrosis, multiple sclerosis, and toxicology. He collaborates on numerous NIH-funded projects involving PCB exposure, microbiome research, and genetic determinants of disease.
  1. Wang, K. (2021). Relating parameters in conditional, marginalized, and marginal logistic models when the mediator is binary. Statistics and Its Interface, 14(2), 109-114.
  2. Wang K (2020) Direct effect and indirect effect on an outcome under nonlinear modeling. The International Journal of Biostatistics 1 (ahead-of-print)
  3. Chen Z, Wang K (2019) Gene-based sequential burden association test. Statistics in medicine 38 (13):2353-2363
  4. Wang K (2019) Maximum likelihood analysis of linear mediation models with treatment-mediator interaction. Psychometrika 84 (3):719–748
  5. Wang K (2018) Understanding Power Anomalies in Mediation Analysis. Psychometrika 83 (2):387-406

Gideon Zamba

  • CV
  • Statistical Methodologies: Gideon Zamba works extensively in change point analysis, sequential analysis, recurrent event modeling, and multivariate statistical process control.
  • Interdisciplinary Applications: His interdisciplinary research spans cancer research, Glaucoma and ophthalmology (including visual field progression modeling), Pulmonary diseases (e.g., emphysema, COPD), Burn injury and trauma, Mental health and psychiatry, Preeclampsia and maternal health, Driving studies and aging, and Global health and development, including mentoring and thesis supervision in Togo.